inst/data-raw/process/PMID30552110_Lam-2019/process.R

In avucoh/nPOD: What the Package Does (One Line, Title Case)

library(readxl)
library(tidyr)
library(dplyr)

stab5 <- read_excel("DB180305SupplementaryData_manual_extraction.xlsx", sheet = "STable5")

# for the final data, we necessarily get consensus values from investigator 1 & 2 for Ins staining assessments
stab5 <- stab5 %>% 
  group_by(ID) %>%
  summarize(strong.ins.prct = mean(c(strong.ins.prct.1, strong.ins.prct.2)), strong.ins.prct_SD = sd(c(strong.ins.prct.1, strong.ins.prct.2)),
            moderate.ins.prct = mean(c(moderate.ins.prct.1, moderate.ins.prct.2)), moderate.ins.prct_SD = sd(c(moderate.ins.prct.1, moderate.ins.prct.2)),
            no.ins.prct = mean(c(no.ins.prct.1, no.ins.prct.2)), no.ins.prct_SD = sd(c(no.ins.prct.1, no.ins.prct.2)))

# we necessarily combine values from investigator 1 & 2 for Syn staining assessments
stab6 <- read_excel("DB180305SupplementaryData_manual_extraction.xlsx", sheet = "STable6") %>%
  group_by(ID) %>%
  summarize(endocrine.prct = endocrine.prct,
            Syn.InsLo.prct = mean(c(Syn.prct.InsLo.1, Syn.prct.InsLo.2)), Syn.InsLo.prct_SD = sd(c(Syn.prct.InsLo.1, Syn.prct.InsLo.2)))
  

# only need to get relative proportions of Ins- and Ins+ islets from given counts
stab7 <- read_excel("DB180305SupplementaryData_manual_extraction.xlsx", sheet = "STable7") %>%
  group_by(ID) %>%
  summarize(InsNeg.islet.prct = round((InsNeg.islets / (InsNeg.islets + InsPos.islets)) *100, digits = 1),
            InsPos.islet.prct = round((InsPos.islets / (InsNeg.islets + InsPos.islets)) *100, digits = 1),
            InsNeg.islet.size = InsNeg.islet.size, InsNeg.islet.size_SEM = InsNeg.islet.size_SEM,
            InsPos.islet.size = InsPos.islet.size, InsPos.islet.size_SEM = InsPos.islet.size_SEM)
  
dataset <- full_join(stab5, stab6, by = "ID") %>% full_join(stab7, by = "ID")
write.table(dataset,"PMID30552110_1_Lam-2019.tsv", sep = "\t", quote = F, row.names = F)